4 ms·

> out yourself as biased garbage? Biased based on your opinion. I personally think this is a very big deal and I don't see anything wrong in that sentence. If

by ccnafr 8y ago

> out yourself as biased garbage? Biased based on your opinion. I personally think this is a very big deal and I don't see anything wrong in that sentence. If you think Google is innocent, just look at the facts. 1) They had an internal memo discouraging public disclosure. 2) They only came clean with the leak after WSJ reached out for comment before the publication of a story. 3) They buried the "leak" announcement in a gigantic blog post. The blog post was meant to announce a new project, but about halfway through the middle they disclosed a data-leaking bug. I'd say the tone of the article is in response to Google's observed and recorded actions. Yeah, I'd say the article and letter's tone is just about right.

kodablah 8y ago
> I personally think this is a very big deal [...] Yeah, I'd say the article[...]'s tone is just about right. That an article needs a "tone", and that it's right/wrong based on how big of a deal the reader thinks it is, is the issue. I await robot news that is a bulleted list of facts. We could argue about fact selection/placement, but I'll take that over arguing a particular organization's collective side of the bed they wake up on. To your point, we disagree there wrt internal discussions of disclosure. What will be real scary is when there is a chilling effect to even performing rational discussion around these things. There is no such thing as disclose all, and when there is no such thing as talking about it either, you'll wish these rules were codified in law instead of applied selectively to issues you are ok with.
- tyrust 8y ago
  >I await robot news that is a bulleted list of facts. Interesting that you mention this, I'm not convinced that this will be a silver bullet. A couple of concerns that come to mind: 1. Something must bridge the robot to the physical world. The selection of which facts to include and exclude in this data source can create bias. 2. From the data source, the robot then would make its own decision about to to present facts, presumably performing some filtration of its own. Such a robot can have bias [0]. To be clear, I definitely agree that robojournalism could be an improvement. But we can't forget how it works and how it could be manipulated. [0] - For example http://blog.conceptnet.io/posts/2017/how-to-make-a-racist-ai-without-really-trying/ http://blog.conceptnet.io/posts/2017/how-to-make-a-racist-ai...
- michaelt 8y ago
  I await robot news that is a bulleted list of facts. I've got just the thing for you: cat news.txt | sed 's/\. /.\n * Reportedly, /'
bilbo0s 8y ago
>Biased based on your opinion... All bias is based on opinion. It's possible for everyone making a comment, including you and I, to be biased. In fact, in any political argument the chances that a commenter is unbiased are vanishingly small. You have your bias. Kodablah has his/her own bias. There's nothing wrong with that. It certainly doesn't negate either of your arguments. Kodablah's point remains just as valid to consider.
Kalium 8y ago
There are two separate items, just in the first sentence quoted, that can reasonably be challenged as unfair. First, there's no evidence to date that the vulnerability in question actually resulted in a data breach. This is distinct from there being actual, affirmative evidence supporting the idea that there was a breach. Here, I am using "vulnerability" to mean there was potential for exploitation and "breach" to mean an actual exploitation and exfiltration of data. I understand that some might opt regard this as a distinction without difference. Second, it is not a common practice to make a loud media-friendly disclosure of every potentially-exploitable vulnerability. I've personally closed dozens of vulnerabilities, potentially exploitable vulnerabilities, in a variety of software packages for a variety of companies over the past decade. Of these, exactly one turned into a public notification of any sort. Hosted service companies in particular are unlikely to create what they would view as an undue media circus around bugs they find and fix internally. With these items in mind, I submit that some people might find it reasonable that the phrasing of the article and the factual claims it makes might be at the very least unfair and potentially possessed of an opportunity to come into greater alignment with evidence-based reality.
- tedivm 8y ago
  Your first point is actually the main problem here- according to Google's internal memo they didn't have the logging data to support a full investigation and have no way of knowing whether a breach occurred or not. If you find a vulnerability and there's no way to prove it hasn't been exploited then the responsible thing to do is treat it as if it had been.
  - Kalium 8y ago
    You're absolutely right! That's very much the responsible thing to do for internal purposes. That said, it might be slightly less than maximally honest for press outlets to claim that the one scenario is factually identical to the other to a public that is ill-equipped to appreciate the point you have so wisely made. At this juncture, I think it's fair to say that Google's internal worries about public response were well-founded. They're being publicly crucified for going above and beyond the common standards of responsible behavior here.

by ccnafr 8y ago

kodablah 8y ago

> I personally think this is a very big deal [...] Yeah, I'd say the article[...]'s tone is just about right. That an article needs a "tone", and that it's right/wrong based on how big of a deal the reader thinks it is, is the issue. I await robot news that is a bulleted list of facts. We could argue about fact selection/placement, but I'll take that over arguing a particular organization's collective side of the bed they wake up on. To your point, we disagree there wrt internal discussions of disclosure. What will be real scary is when there is a chilling effect to even performing rational discussion around these things. There is no such thing as disclose all, and when there is no such thing as talking about it either, you'll wish these rules were codified in law instead of applied selectively to issues you are ok with.

tyrust 8y ago
>I await robot news that is a bulleted list of facts. Interesting that you mention this, I'm not convinced that this will be a silver bullet. A couple of concerns that come to mind: 1. Something must bridge the robot to the physical world. The selection of which facts to include and exclude in this data source can create bias. 2. From the data source, the robot then would make its own decision about to to present facts, presumably performing some filtration of its own. Such a robot can have bias [0]. To be clear, I definitely agree that robojournalism could be an improvement. But we can't forget how it works and how it could be manipulated. [0] - For example http://blog.conceptnet.io/posts/2017/how-to-make-a-racist-ai-without-really-trying/ http://blog.conceptnet.io/posts/2017/how-to-make-a-racist-ai...
michaelt 8y ago
I await robot news that is a bulleted list of facts. I've got just the thing for you: cat news.txt | sed 's/\. /.\n * Reportedly, /'

bilbo0s 8y ago

>Biased based on your opinion... All bias is based on opinion. It's possible for everyone making a comment, including you and I, to be biased. In fact, in any political argument the chances that a commenter is unbiased are vanishingly small. You have your bias. Kodablah has his/her own bias. There's nothing wrong with that. It certainly doesn't negate either of your arguments. Kodablah's point remains just as valid to consider.

Kalium 8y ago

There are two separate items, just in the first sentence quoted, that can reasonably be challenged as unfair. First, there's no evidence to date that the vulnerability in question actually resulted in a data breach. This is distinct from there being actual, affirmative evidence supporting the idea that there was a breach. Here, I am using "vulnerability" to mean there was potential for exploitation and "breach" to mean an actual exploitation and exfiltration of data. I understand that some might opt regard this as a distinction without difference. Second, it is not a common practice to make a loud media-friendly disclosure of every potentially-exploitable vulnerability. I've personally closed dozens of vulnerabilities, potentially exploitable vulnerabilities, in a variety of software packages for a variety of companies over the past decade. Of these, exactly one turned into a public notification of any sort. Hosted service companies in particular are unlikely to create what they would view as an undue media circus around bugs they find and fix internally. With these items in mind, I submit that some people might find it reasonable that the phrasing of the article and the factual claims it makes might be at the very least unfair and potentially possessed of an opportunity to come into greater alignment with evidence-based reality.

tedivm 8y ago
Your first point is actually the main problem here- according to Google's internal memo they didn't have the logging data to support a full investigation and have no way of knowing whether a breach occurred or not. If you find a vulnerability and there's no way to prove it hasn't been exploited then the responsible thing to do is treat it as if it had been.
- Kalium 8y ago
  You're absolutely right! That's very much the responsible thing to do for internal purposes. That said, it might be slightly less than maximally honest for press outlets to claim that the one scenario is factually identical to the other to a public that is ill-equipped to appreciate the point you have so wisely made. At this juncture, I think it's fair to say that Google's internal worries about public response were well-founded. They're being publicly crucified for going above and beyond the common standards of responsible behavior here.